1-'1'>I McEiri n: Iltidroiinipliic ()hs( rnitinns of Scripps I iislil ill inn 2.19 



always the effect due to the much slower processes of conductiou and 

 diffusion, both of which tend to produce a state of homogeneity ver- 

 tically as well as horizontally. While any type of circulation which 

 mixes the water accelerates these equalizing processes, there are many 

 factors, such as variation in solar radiation, condition of the air, etc., 

 that have the opposite effect. Accordingly, the distribution of the 

 various properties of sea-water may be exceedingly complex. 



Finally, while the distribution of any one property of ocean water 

 affords indirect information concerning its circulation, the combined 

 evidence from more than one is still more valuable. Moreover, in 

 some cases the distribution of sevtM-al properties, sueli as temperature, 

 salinity, and gas content, must be known in order to work out the 

 circulation satisfactorily. 



(6) Methods employed in detennining isutherms and isolialines. — 

 To determine the exact distribution of temperatures or salinities in 

 a given area at a given time requires a large number of simultaneous 

 observations well distributed throughout the area. On the basis of 

 these observations lines of equal temperature or salinity can be drawn 

 to represent the distribution at that time. All of our observations 

 were made from one boat, and therefore could not be simultaneous. 

 Moreover, they were not as well distribiited as we wished, except in 

 a few portions of the area. Nevertheless, an approximation to tlie 

 average distribution of temperature and salinity during certain months 

 could be worked out in some cases by averaging and interpolating. 

 This has been done, and after applying a correction for the diurnal 

 variation (see p. 271) the average of the values obtained during the 

 particular month considered was computed for each section and was 

 assumed to correspond to the position of its center. It is well to 

 point out, therefore, that the effect of only those approximately con- 

 tinuous influences having, on the average, a tendency to produce fairly 

 constant types of temperature and salinity distribution dui'ing the 

 periods of observation are likely to be revealed by the isotherms or 

 isohalines determined in this waj'. 



(c) Isotherms. — The southward dip of the isotherms southeast of 

 Santa Catalina Island (pis. 5 and (i) and southeast of San Clemente 

 Island (pis. 7 and 8, figs. 7 and 8) reveals a southward flow of water, 

 since lower temperatures are found in the vicinity of the portions of 

 greatest curvature than either east or west of them. The source of this 

 cold water will be considered later. Furthermore, the southerly drift 

 thus revealed agrees, in general, with the average of direct current 



